The proposed research project will focus on the difficulties children with ASD have understanding and participating in conversations. These difficulties are often attributed to global and poorly defined deficits in Theory of Mind, the abilty to infer mental states or perspectives of others, Weak Central Coherence, understanding global contexts, and Executive Function, cognitive control. Unfortunately these theories and much of the literature focus on nonverbal interpersonal actions. The proposed project focuses on scalar implicature, conversational inferences when an utterance has a scalar (pair of words reflecting scale ends). Some students danced: the scalar term some can be interpreted either lexically (i.e., up to and including all students) or pragmatically (i.e., up to but not including all studens). An implicature is made for the pragmatic interpretation of the weaker scalar. This process, an interface between semantics and pragmatics, provides a more specific way of examining pragmatic deficits. The ability to derive implicatures in language is affected by two variables: cognitive effect and effort. Cognitive effect is the communicative value a listener places on computing an implicature; anticipating enriched meaning leads to implicatures. Cognitive effort, more accurately processing costs, can be affected by working memory. Data will come from 40 children with and without ASD from several sources: standardized tests [NVIQ (TONI - 4), adaptive behavior (VABS - II, expressive and receptive language (CELF - 4), and vocabulary (PPVT - IV)] and three experimental eye tracking studies. The goal of these studies is to examine the comprehension of scalars (some/all; or/and) and cognitive effects and efforts between and within diagnostic groups. Eye tracking will allow us to examine the process of decision-making through eye gaze which is a more subtle measure than reaction time or accuracy. Exp. 1 and 3 are 4AFC (match, scalar alternative, semantic alternative, and unrelated) sentence-picture matching tasks. Exp. 1 examines cognitive effect (informativeness). Exp. 3 also examines cognitive effort (grammatical strengthening). Exp. 2 is a 2AFC sentence-picture judgment task comparing computerized versions of Truth Value Judgment Task (high cognitive effort) with Felicity Judgment Task (low cognitive effort). Children, 7.0 - 8.4 years of age, with ASD will be selected by ADOS scores and NVIQ (WNL). This HFA group will be compared to gender and age- or language-matched (based on clinical group scores) children without ASD. Results on standardized testing will be compared to accuracy, reaction time, and eye tracking data to assess their roles in scalar implicature derivation. The proposed project will add to the limited body of research on specific linguistic abilities of school- aged children with autism, may lead to a more specific understanding of language and cognitive deficits in autism, and add to our knowledge of the general nature of these disorders by informing the current theories of ASD. It may also have assessment and intervention implications for these children.